A quick MOSS-Image news round-up - 19th October 2026
retained text
Another quick MOSS news and goodies round-up, for those who may have missed some items. -> ComfyUI Portable is now officially at v0.42.1, with the .1 increment being to... "Fix some crashing issues a few users experienced when running the MOSS image template workflow." https://github.com/Comfy-Org/ComfyUI/releases -> MOSS-Layout-Editor. Draw boxes, move boxes, nudge boxes into place, label boxes with region-specific prompts and/or the text to render. https://github.com/spooky-local/MOSS-Layout-Editor -> ComfyUI-MOSSLayout. Another visual layout editor, this one... "forked from ComfyUI-IdeogramHelper". https://github.com/0Printemps0/ComfyUI-MOSSLayout -> MOSS-Image as a line-art colouriser. The input image goes in via... "Qwen3.5's vision tower inside the text encoder, and as a clean VAE latent in the DiT sequence. The output keeps the page's line work, faces and lettering." LoRA, workflow and img2img node available. The required ComfyUI nodes appear to be a way to do general Img2Img with MOSS? https://huggingface.co/sorryhyun/moss-image-colorize https://github.com/sorryhyun/ComfyUI-MOSS-i2i (required Img2Img nodes) -> MOSS-Image Color-Depth Control. "A control adapter for MOSS-Image that makes generation follow a color region map (where each object is) and a depth map (its 3D shape and front/back order)". Designed for something called Hayate, which I've never heard of - and which is '404 not found' on following the given HF link. No ComfyUI implementation or workflow, as yet. https://huggingface.co/darask0/MOSS-Image-color-depth-control -> A 'model patch' for MOSS-Image Color & Depth Control. Regrettably the required ComfyUI workflow is trapped on civitai .red which I can't access from the UK even with a VPN. Still, this and the previous item suggest that a combo of colour and depth control inputs can act as a basic Controlnet for MOSS-Image. https://civitai.com/models/3010285/moss-color-and-depth-control -> Basic sliders for lighting and age for MOSS-Image. https://civitai.com/models/3010015/moss-lighting-slider https://huggingface.co/Playtime-AI/Moss-Image-Age_Slider/tree/main -> And finally, Fizgig 8.7 has tweaked MOSS-Image LoRA training, for better and more consistent likenesses. https://old.reddit.com/r/StableDiffusion/comments/1x9khtm/a_potential_fix_for_moss_training_issues/ ~ OLD POSTS ~ https://old.reddit.com/r/comfyui/comments/1x97q81/a_quick_mossimage_news_roundup_18th_october_2026/
document fetched 2026-10-20 00:41 · canonical https://reddit.com/r/comfyui/comments/1xa3yk2/a_quick_mossimage_news_roundup_19th_october_2026